All Products
Search

This Product

    :Configure task scheduling

    Document Center

    :Configure task scheduling

    Last Updated:Oct 08, 2026

    This topic uses the write_result task, configured to run weekly, to show how to set scheduling and dependency properties in DataWorks.

    Prerequisites

    • You have created a zero-load node (start) and an ODPS SQL node (insert_data). For more information, see Create a workflow.

    • You have created a synchronization task from the synchronization node write_result. For more information, see Create a synchronization task.

    Background

    DataWorks supports scheduling triggered by time and dependencies. It can schedule tens of millions of tasks each day accurately and on time according to their DAG relationships. You can configure minute, hour, day, week, or month scheduling cycles. For more information, see Configure time properties.

    Configure time properties

    1. Go to the DataStudio page.

      Log on to the DataWorks console. In the target region, click Data Development and O&M > Data Development in the left-side navigation pane. Select a workspace from the drop-down list and click Go to Data Development.

    2. In the corresponding workflow, double-click the offline synchronization node write_result to open its editing page.

    3. In the navigation pane on the right side of the editing page, click Scheduling.

      Note

      Nodes created in a manual workflow must be triggered manually and cannot run through scheduling.

    4. In the Time Properties area, configure the task's scheduling time properties.

      Parameter

      Description

      Instance Generation Mode

      Options include Next Day and Immediately After Deployment. For more information, see Instance generation modes.

      Recurrence

      • Normal: Starts scheduling at the configured time in each scheduling cycle and runs the task normally to process data.

      • Suspend scheduling: Starts scheduling at the configured time in each scheduling cycle, but pauses the node so that it does not process data.

      • Dry Run: Starts scheduling at the configured time in each scheduling cycle, but runs the node as a zero-load task without processing data.

      Scheduling period

      The node's run cycle can be yearly, monthly, weekly, daily, hourly, or by minute. In this example, scheduling starts at 00:00 every Monday and Tuesday.

      cron expression

      Displayed by default based on the configured scheduling time and cannot be changed.

      Timeout Definition

      The task is automatically terminated if its run duration exceeds the timeout.

      • The timeout applies to recurring instances, backfill instances, and test instances.

      • The default timeout is dynamically adjusted based on actual system load and ranges from 3 to 7 days.

      Note

      • Timeout DefinitionThe maximum value can be set to 168 hours.

      • Traffic, computing, and other charges incurred by a task still apply if the task is terminated because its execution exceeds the timeout.

      Rerun attribute

      Options include You can run again after successful or failed operation., Do not re-run after successful operation, and re-run after failure. and Disallow Regardless of Running Status.

      Auto Rerun upon Failure

      If Rerun attribute is set to You can run again after successful or failed operation. and Do not re-run after successful operation, and re-run after failure., this property is displayed and you can configure automatic reruns after task errors. If it is set to Disallow Regardless of Running Status, this property is hidden and the task will not rerun automatically after an error.

      Rerun Times

      When you select Error automatic re-run, you must configure Rerun Times.

      Rerun Interval

      When you select Error automatic re-run, you must configure Rerun Interval. The default interval between reruns is 30 minutes. You can set it to a minimum of 1 minute and a maximum of 30 minutes.

      Validity Period

      The node's validity period. Set it based on your requirements.

      For more information about time properties, see Configure time properties.

    Configure scheduling dependencies

    Dependency properties specify the node's upstream dependencies. Even when the current node's instance reaches its scheduled time, it starts only after the upstream instances finish running.

    For example, the current node's instance starts only after the upstream insert_data node's instance finishes running.

    Note

    In business terms, a downstream node must wait for an upstream node to produce table data before it can process that data, such as cleaning the upstream table or writing the cleaned results back to another database. These operations require the upstream node to run successfully and produce the data. Scheduling dependencies ensure that the upstream data is available when the downstream node starts. For more information about scheduling dependencies, see Scheduling dependency configuration guide.

    By default, the scheduling system creates a workspace_name_root node as the root of each workspace. A node with no other upstream node can depend directly on this root node.

    Submit and publish the task

    1. In the write_result node's editing page, click the Save icon in the toolbar to save the node.

    2. Submit the task.

      Note

      You must set the node's Rerun attribute and Parent Nodes before you can submit the task.

      1. Click the Submit icon in the toolbar.

      2. In the Submit dialog box, enter Change Description.

      3. Click OK.

      If you use a Standard mode workspace, the synchronization task is available in the development environment after submission. To publish it to Production Environment for periodic scheduling, click Task Publishing. Publish the synchronization task to the production environment on that page. For instructions, see Publish tasks.

      A synchronization task starts generating instances at the configured times and running automatically according to its scheduling cycle from the next day only after it is submitted to the scheduling system.

      Note

      If you submit the synchronization task after 23:30, the scheduling system starts generating recurring instances and running them automatically on the third day.

    Next steps

    You have learned how to configure scheduling and dependency properties for an offline synchronization task. In the next tutorial, you will learn how to perform recurring task operations and inspect logs to troubleshoot the submitted task. For more information, see Run and troubleshoot tasks.